{"ID":2888757,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.22576","arxiv_id":"2507.22576","title":"COOkeD: Ensemble-based OOD detection in the era of zero-shot CLIP","abstract":"Out-of-distribution (OOD) detection is an important building block in trustworthy image recognition systems as unknown classes may arise at test-time. OOD detection methods typically revolve around a single classifier, leading to a split in the research field between the classical supervised setting (e.g. ResNet18 classifier trained on CIFAR100) vs. the zero-shot setting (class names fed as prompts to CLIP). In both cases, an overarching challenge is that the OOD detection performance is implicitly constrained by the classifier's capabilities on in-distribution (ID) data. In this work, we show that given a little open-mindedness from both ends, remarkable OOD detection can be achieved by instead creating a heterogeneous ensemble - COOkeD combines the predictions of a closed-world classifier trained end-to-end on a specific dataset, a zero-shot CLIP classifier, and a linear probe classifier trained on CLIP image features. While bulky at first sight, this approach is modular, post-hoc and leverages the availability of pre-trained VLMs, thus introduces little overhead compared to training a single standard classifier. We evaluate COOkeD on popular CIFAR100 and ImageNet benchmarks, but also consider more challenging, realistic settings ranging from training-time label noise, to test-time covariate shift, to zero-shot shift which has been previously overlooked. Despite its simplicity, COOkeD achieves state-of-the-art performance and greater robustness compared to both classical and CLIP-based OOD detection methods. Code is available at https://github.com/glhr/COOkeD","short_abstract":"Out-of-distribution (OOD) detection is an important building block in trustworthy image recognition systems as unknown classes may arise at test-time. OOD detection methods typically revolve around a single classifier, leading to a split in the research field between the classical supervised setting (e.g. ResNet18 clas...","url_abs":"https://arxiv.org/abs/2507.22576","url_pdf":"https://arxiv.org/pdf/2507.22576v1","authors":"[\"Galadrielle Humblot-Renaux\",\"Gianni Franchi\",\"Sergio Escalera\",\"Thomas B. Moeslund\"]","published":"2025-07-30T11:02:38Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\",\"cs.LG\"]","methods":"[]","has_code":false,"code_links":[{"ID":611575,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2888757,"paper_url":"https://arxiv.org/abs/2507.22576","paper_title":"COOkeD: Ensemble-based OOD detection in the era of zero-shot CLIP","repo_url":"https://github.com/glhr/COOkeD","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}